Postdoc (m/f/d) Multi-task deep learning for multimodal image analysis

Beginn der Ausschreibung11.02.2020
Ende der Ausschreibung10.03.2020
InstitutInstitut für Werkstoffforschung
KontaktpersonErika Krüger

Research at HZG in the field of “Metallic Biomaterials” focusses on the development of biodegradable Mg-based alloys and implants by the optimization of mechanical, biological and degradation properties for specific applications. To this end, it is crucial to understand the mechanisms governing the interactions between microstructure, mechanical properties, biology and degradation. In the long run, such an understanding will enable the manufacturing of implants tailored to the specific needs of individual patients. To achieve this, large amounts of multimodal biomedical image data (laboratory / synchrotron CT, MRI, histology, SAXS etc.) have been acquired which have to be processed and analyzed.

In order to process and analyze these multimodal data, you will develop a unified framework that enables a quick and efficient implementation and application of common image processing tasks. Within this project, we focus on the tasks of image enhancement (reduction of artefacts, noise, etc.), segmentation and registration. Multi-task learning solutions will be employed in order to simultaneously address these tasks and to exploit synergetic effects. The unified framework shall facilitate the application of these multi-task solutions across domains and modalities.The position will initially be limited to three years. The place of employment is Hamburg (outstation at DESY).

Your tasks:

  • development of multi-task learning solution for image enhancement, segmentation and registration
  • development of a unified framework for multi-task analysis of multimodal data
  • close collaboration with project partners to ensure knowledge transfer regarding image processing tasks
  • deployment of algorithms and framework at the HPC cluster at PETRA III at DESY
  • performance evaluation of algorithms and framework
  • application of algorithms and framework on biomedical data (and other)
  • development of a common data format or solutions to facilitate access and exchange of data between partners
  • preprocessing and databank integration of multimodal data sets
  • project management tasks
  • publication of scientific results in international scientific journals & presentation of scientific results at international conferences and workshops

Your profile:

  • PhD in Physics, Mathematics, Scientific Computer Science or in a related discipline
  • strong background in deep learning and software/algorithm development
  • experience in imaging
  • experience in multi-modal data analysis
  • experience in multi-task learning
  • working knowledge of image processing (segmentation, image enhancement, registration)
  • working experience in common machine learning libraries such as TensorFlow or PyTorch
  • very good programming skills in a high level computer language
  • excellent oral and written English skills

For further information please contact:

Dr. Julian Moosmann

We offer you:

  • multinational work environment with over 1000 colleagues from more than 50 nations
  • extensive options of vocational training (i. a. expert seminars, language courses or leadership seminars)
  • flexible working hours and various models to ensure ensure the compatibility of family and career
  • excellent infrastructure including modern work spaces
  • remuneration according to the standards of the collective wage agreement TV-AVH including further social benefits

The promotion of equal rights is a matter of course for us. Severely disabled persons and those equaling severely disabled persons who are equally suitable for the position will be considered preferentially within the framework of legal requirements.

Then we are looking forward to receiving your comprehensive application documents (cover letter, CV, transcripts, certificates etc.) indicating the reference number code no. 2020/WB 1. Closing date for applications is March 10th, 2020.

Please send your application to:

human resources